Vikash Kaushik

Your boy is in krabi town 😊

@kubesimplify Yeah definitely.

We are planning a free KubeSimplify guide: Day-2 ops for GPU clusters. 30 pages. No email gate. No paywall. Inside: — Driver upgrade playbook — DRA migration runbook — Capacity planning spreadsheet — Incident response template Let us know if it sounds useful for your team.

@vedantj_03 agla match to khilana hi chaiye 🤌

Dhoni not playing today

@cloclodma Rightly says - DevOps :)

Politicians say: If you want peace, prepare for war. DevOps say: If you want to be taken seriously, master Kubernetes.

@livingdevops I have seen this in TVs :), I haven't worked from the office yet But this is good.

How you rate this work desk setup?

@livingdevops I should try it then

Codex > Claude Code? Can I move my $100/month from claude to codex?

@brankopetric00 And we rarely use it, but we pay for it monthly: (

AWS Business Support costs more per month than rent in most cities

@codoyevskyy I have to say this is the fantastic breakdown 🙌

so if you have 3 node cluster, it can tolerate 1 node failure as it has quorum = 2, while if you have 4 node cluster it will also tolerate 1 node failure having quorum 3. you see 1 node is getting wasted in even numbered (4/4)

so in a k8s cluster etcd should have odd replicas to help avoid split brain conditions because this is where replicas experience communication issues and cannot be sure if they have a quorum now....... 🧵(1/4)

@codoyevskyy No worries, surey next time :) I had gotton the same mail in my college years : ) Since then attended 2 kubecon and this one will be 3rd one

maybe next time

@cloclodma If the backend volume expanded but df -h still shows 10Gi, restart the pod or trigger filesystem resize depending on the storage driver and filesystem type.

@cloclodma I’d start by checking the PVC and StorageClass. Updating the PVC request alone does not guarantee the filesystem inside the pod is resized. I’d verify that allowVolumeExpansion is enabled, the CSI driver supports expansion, and whether the PVC is stuck in FileSystemResizePending

You deployed a database in Kubernetes with a PVC of 10Gi. A few days later, the app starts failing with: No space left on device You update the PVC to 30Gi, but inside the pod, df -h still shows 10Gi. Where would you start investigating?

One concern I see with AI is that engineers may miss small details, which can lead to serious credential leaks and more security incidents at major IT companies.

@asynctrix Nice and onpoint breakdown

DevOps Tools and their Difficulty to Learn 🔥 • 📋 YAML → 🟢 Easy • 🐙 Git → 🟢 Easy • 🐧 Linux CLI → 🟢 Easy • 📜 Bash Scripting → 🟡 Easy–Medium • 🐳 Docker → 🟡 Easy–Medium • 🔄 GitHub Actions → 🟡 Easy–Medium • 📦 Ansible → 🟡 Easy–Medium • 🏗️ Terraform → 🟠 Medium • ⚙️ Jenkins → 🟠 Medium • ☸️ Kubernetes → 🟠 Medium • 📊 Prometheus + Grafana → 🟠 Medium • ☁️ AWS/GCP/Azure Basics → 🟠 Medium • 🛠️ Helm → 🟠 Medium • 🌐 Advanced Kubernetes (Networking/RBAC) → 🔴 Hard • 🔄 GitOps (ArgoCD/Flux) → 🔴 Hard • 🛡️ Security & IAM → 🔴 Hard • 📡 Full Observability Stack → 🔴 Hard • 🌐 Istio Service Mesh → 🟣 Very Hard • 🧬 Custom Kubernetes Operators → 🟣 Very Hard • ⚡ eBPF / Cilium → ☠️ Extreme • 🏭 Building Your Own Orchestrator → ☠️ Extreme

CI failures halt progress. Feeding detailed build logs to an LLM like Claude 3 can quickly pinpoint issues. Prompt it for error patterns or root causes. This bypasses tedious manual log grep, significantly cutting debug time. #CI/CD #AI

learn.microsoft.com/en-us/azure/ak… #Azure #AKS

AKS cluster costs sneak up with over-provisioned nodes and inefficient autoscaling. Resource requests for Pods are key. How do you fine-tune AKS autoscaling for cost efficiency without sacrificing performance?

Which LLM do you find most effective for generating or validating Infrastructure as Code?

gonna deploy loki and thinking to integrate to send it to slack with RCA

deployed prometheus mcp very useful for devlopers.

How do your teams primarily access runbooks for Kubernetes incidents?

Stale runbooks prolong incident resolution dramatically. Reliable incident response needs accurate, maintained guides. What version control or automation do you use for your incident response documentation? github.com/github/docs #SRE #DevOps